Multisampling Model Predictive Control For PMSM Drives With Improved Tracking Performance and High Bandwidth

Model predictive control (MPC) has received great favor in AC motor drive applications with fast dynamic benefits. However, the MPC is also vulnerable to inherent digital delay, parameter mismatch, and unknown disturbance, resulting in inaccurate dq-axes current tracking performance, especially under low carrier frequency. The multisampling technique introduces multiple available state feedback into the control loop, which has been proven to be a powerful strategy in terms of control delay reduction and bandwidth enhancement. However, rare research has combined these two techniques to preserve their benefits simultaneously and there also lacks a deeper analysis of MPC under the multisampling mode. To fill this research gap, this paper analyzes the control law generation and tracking performance of the multisampling MPC (MS-MPC). The experimental results show that the bandwidth can be achieved as high as possible and the controller’s robustness can be enhanced compared to conventional MPC.